Infinex Litepaper

Version 0.1

Abstract

The Infinex platform provides access to onchain protocols, services and applications. It is non-custodial, using a novel security architecture centred around onchain smart accounts and passkeys¹. Infinex currently supports six EVM and non-EVM chains, including Solana, Base and Arbitrum. Infinex is designed to replace centralized platforms as the primary point of interaction for crypto users.

Ecosystem Overview

Centralized exchanges have dominated the crypto ecosystem for over a decade, beginning with Mt Gox. There is some irony in a peer-to-peer electronic cash system relying so heavily on centralized platforms for distribution. This situation arose because most users opt for convenience over security. Centralised platforms also enable more efficient speculation; early adoption was much faster because exchanges aggregated liquidity far better than peer-to-peer exchanges. Bitcoin adoption would, therefore, likely have been much slower without these centralized services. While there is nothing inherently wrong with bootstrapping network effects via centralization, the cost to users of these centralized services has been very high due to hacks and fraud.

We no longer need to compromise. We have reached an inflection point where decentralized infrastructure can compete with and exceed centralized services. It is time for a new phase of growth.

Core Philosophy

Infinex is built to optimise the User Experience. To supplant centralized services, Infinex must replicate the full user experience of centralized platforms. This contrasts with most decentralized services that start with infrastructure and protocol design and build out to the UX layer. For too long, we have built parts of the experience and expected or hoped adoption would follow. Empirical evidence² has completely disabused us of this notion. However, while replicating the UX of centralized services, Infinex must always remain non-custodial. Within these two constraints, many pragmatic decisions have been and will continue to be made to achieve the outcome of replacing centralized exchanges and other custodial services.

Feature Overview

Centralized services offer many features, which are listed and described below. Many different protocols can deliver these features. The following list is indicative only and represents protocols that have either been discussed by the community or have already passed governance proposals. The Infinex governance process is described later in this paper, but it is open, so anyone can propose integrating a new chain or protocol to deliver new functionality to Infinex users.

Custody

Storing assets safely is the primary use case of Centralized platforms; however, to do so, they require the user to hand over custody. This can help users avoid losing access to their crypto, but it puts the user at risk of the platform stealing or losing their funds, either through hacks, fraud or negligence.

The primary custody solution used by the Infinex platform consists of a combination of passkeys, Turnkey, and individual smart accounts deployed to each chain. A proposed extension uses NEAR chain signatures³ to deploy addresses to legacy chains like Bitcoin and Litecoin, allowing for self-custody while retaining the ability to recover from a lost passkey or other failure modes.

Spot Trading/Swaps

Exchanges were created to provide a venue for users to swap or trade between different assets. Initially, this was between fiat and Bitcoin, but many different assets, including stablecoins, were eventually added. The complexity of supporting swaps grew further as new blockchains emerged in the early years of crypto. Today, one of the primary use cases of centralized exchanges is enabling users to trade assets across different blockchains.

Infinex currently supports six chains with significant onchain liquidity across many different liquidity pools. The first swap implementation on EVM chains was Curve Finance⁴, which allows swaps between many different assets with stable ratios. Most Curve liquidity is within Stablecoin pools, but other pools exist for various implementations of volatile assets like ETH and LST/LRT tokens. The first swap implementation for Solana will be Jupiter Exchange [Insert XIP Link 5 TBC].

Margin

Borrowing to trade spot assets is a key feature of most trading venues. However, the use of margin for spot trading has declined significantly over the years as perpetual futures have grown in adoption. Interestingly, though, permissionless lending enables margin trading, and much of the demand for loans onchain is for this purpose.

Several protocols enable margin trading directly onchain on both EVM and Solana. However, initially, margin trading will likely be enabled by allowing users to borrow stablecoins against their volatile assets and then swap into a new (or the same) volatile asset.

Lending

The last cycle saw the rapid emergence and collapse of centralized lending services. The reasons for this growth and collapse are beyond the scope of this paper, but the primary issue was a misalignment of incentives due to the principal-agent problem. CeFi lenders were incentivized to maximise interest return, which led to a race to the bottom, where they lent to the highest-risk counterparties with little regard for preserving collateral. The collapse of platforms like Celsius and Blockfi clearly highlighted the incentive misalignment that can emerge when users give up custody of their assets.

Lending protocols like Aave⁵ are extremely robust and secure and have many years and billions of dollars worth of assets secured across numerous chains. For EVM this will likely be the starting point for borrowing and lending. However, several new protocols offer different mechanisms and trade-offs that should also be explored, including Ammalgam⁶ and Morpho⁷. For Solana, Kamino⁸ is likely the first lending integration, although other novel projects are being developed that leverage the faster blocktimes of Solana.

Perps

Crypto Perpetual Futures were invented by some clever young gentlemen named A. Hayes, B. Delo, and S. Reed. Perps unlocked significant liquidity in high-cap assets like Bitcoin, Litecoin, and Ethereum. Eventually, most centralized platforms rolled out similar features, with FTX extending these markets significantly prior to their collapse in 2022.

The origin of Infinex was an exploration of the lack of adoption of decentralized perps protocols, including Synthetix⁹, dYdX¹⁰, GMX¹¹, and Drift¹². All of these are options for integration, however, due to familiarity it is likely that the focus will be on Drift and Synthetix in the short term.

Options

Options, while extremely popular in TradFi, have experienced less adoption in crypto. Numerous theories surround this, including perps being an easier and more efficient mechanism for leveraged trading. The volume differential between centralized options platforms in crypto and decentralised platforms is greater than for any other service. Whether integrating options into a more user-friendly interface will drive adoption and close this gap remains to be seen.

The most liquid onchain options protocol today is Derive¹³, and again due to familiarity, it is likely this will be the first integration of options. However, there are many other options, like instruments that have been deployed over the years; these include Squeeth¹⁴ by Opyn¹⁵ and numerous tokens deployed by UMA¹⁶.

Staking

While staking has existed for almost as long as crypto, the transition to PoS for Ethereum significantly accelerated this trend. Most centralized exchanges have some form of custodial staking service available for several different assets and blockchains, including Solana and Ethereum.

The two largest staking ecosystems are Solana and Ethereum, the primary mechanism for enabling staking is from LSTs (Liquid staking tokens). Infinex already supports many LRT tokens, including jitoSOL and wstETH. Eventually, the ability to stake SOL and ETH on the platform into liquid staking protocols will likely be added.

NFTs

NFTs exploded in 2021 after a long winter following their brief adoption in late 2017 on Ethereum. NFTs have been around for almost as long as centralized exchanges and yet no centralized platform managed to integrate them successfully. Even today the dominant trading venues for NFTs are onchain. This presents an opportunity for a platform that can safely store and trade NFTs across multiple blockchains.

Blur¹⁷ is the most liquid trading venue for NFTs currently, so this is likely to be the first trading integration for NFTs, however, prior to this integration the ability to custody NFTs will be enabled.

Launchpads

Binance¹⁸ initiated the IEO following the collapse of the ICO boom in 2018. The concept was to launch new projects directly with an exchange listing. While these platforms have pivoted away from direct token launches over the years, there is a significant opportunity for a decentralised solution.

The rise of platforms like pump.fun¹⁹ has been significant over the last year. However, these platforms have high barriers to entry for centralized exchange users. By enabling easier and safer onboarding allowing users to access earlier stage tokens and memecoin will drive significant adoption.

Yield farming

One of the directions that launchpads have gone in is to offer staking to earn tokens in new projects; this is an attempt to replicate airdrop and yield farming that is more commonly executed onchain. One of the key advantages of a platform that uses onchain infrastructure is to be able to directly access these kinds of bootstrapping initiatives, whether they be offered by blockchain protocols or other projects.

Every new network and platform uses growth incentives to drive awareness and usage. While centralized exchanges attempt to replicate this process it is highly inefficient and less effective in driving user behaviour. Infinex can provide significant user value and differentiation by connecting users directly to these incentives.

User Acquisition

Crypto adoption is extremely cyclical, which means that at the peak of each adoption cycle, the vast majority of users are new entrants into the ecosystem. Each cycle has seen growth and adoption increase by at least an order of magnitude. This presents a significant opportunity for any new platform attempting to challenge the incumbents from the previous cycle. This is evident when looking at the dominant platforms across each wave of adoption; when looking at dominance, we also have empirical external evidence. While there is noise in this data, by observing trends across multiple data sets, we can form a fairly clear picture of which platforms had dominant market share at a given time. For simplicity, we are excluding regional exchanges and focussing on only exchanges that serve(d) a global audience.

Cycle	ATH (BTC)	ATH Date	Dominant CEX
2009 - 2011	$29.58	09-06-2011	Mt Gox
2012 - 2015 ²⁰	$1242	29-11-2013	Mt Gox (pre-collapse), Bitstamp [²¹, ²²]
2015 - 2018	$19783	18-12-2017	Binance, Bitmex (Perps)
2019 - 2022	$67566	11-08-2021	FTX, Binance
2023 - TBC	$73700	13-03-2024	Binance ²³

This list shows a clear pattern of challenger platforms' success in displacing incumbents. This pattern has two primary drivers: the sheer number of new users entering the space with minimal brand awareness or allegiance and the shift in product focus across cycles. Across the five cycles so far in the crypto ecosystem, there have been multiple drivers of new user acquisition. Early on it was Bitcoin demand and the rise of altcoins. Later the emergence of ICOs, perpetual futures with 100x leverage and long tail perps drove the next wave of adoption. More recently the emergence of the Solana ecosystem. This ignores the rapid rise and decline in NFT demand, which occurred primarily on decentralised platforms like Opensea²⁴ and Blur. While it is unclear what will drive user demand and attention during this cycle, Infinex is well-positioned to take advantage of the many emerging applications and services being built onchain. Growth campaigns for onchain services are primarily driven by token and points incentive schemes; this will likely be a powerful acquisition strategy for Infinex because the platform will be best positioned to provide safe and simple access to these campaigns.

Distribution

Distribution is the power to capture and direct user attention. Crypto distribution has been dominated by centralised platforms, except for a brief period during NFT adoption. This is problematic for several reasons. The primary reason is that centralized services do very little for and often undermine the security of blockchains. Take Bitcoin: the more centralized services and custody providers dominate the ecosystem the fewer transactions settle on the blockchain. Peer-to-peer activity drives value for the Bitcoin blockchain, which is critical for security as block space demand is directly correlated to the network's security; this is increasingly true as block rewards decline on a Bitcoin basis each halving cycle. The second reason is that exchanges can only focus distribution power on asset prices, distorting markets and undermining the fundamental assessment of networks and protocols. When listing on centralized trading venues is more important than the launch of a network for attention and awareness, incentives are severely distorted.

The market is severely distorted today, but by adopting a platform that is facilitated by the technology that is being traded, a powerful feedback loop will emerge where the protocols and applications that are most efficient capture increasing market share and awareness; this will restore fundamental market dynamics that ensure strong competition and efficiency across the ecosystem.

A platform that runs on the infrastructure onchain projects develops is far healthier for the ecosystem. Infinex at scale will have the power to drive fundamental usage of protocols and onchain services rather than narrowly focusing on asset prices alone. This should help create a much healthier alignment within the ecosystem as each platform and application's fundamental utility and efficiencies can be directly compared due to a single platform integrating and aggregating them all.

Technical Architecture

Traditionally, crypto protocols have been deployed to one or more chains with separate front-end web clients that connect via an RPC; users interact with the protocol using a browser-integrated wallet. This architecture is brittle because there is no state stored in the client. All state is stored onchain, which leads to poor UX. This architecture ensured users were maximally reliant on onchain components rather than interfaces. The issue is that as protocols become more complex, the ability to integrate a protocol into an interface increases significantly, so while this architecture was ostensibly designed to maximise censorship resistance, it instead created a fragile point of failure at the client layer. In addition, this architecture is fundamentally limited in how well it can optimise for usability; we hit this limit years ago.

Infinex takes a different approach; it maximises censorship resistance at the protocol layer but then optimistically assumes the rest of the stack will remain online. While this approach is somewhat antithetical to the cypherpunk ethos, it is the optimal posture for a platform designed for mass adoption. Thin clients and the poor UX they deliver have pushed millions of people onto centralised platforms. By bridging the UX gap and accepting that censorship resistance should not be the driving factor in platform design, we can deliver a user experience that migrates users away from centralised platforms.

The Infinex Platform is a vertically integrated app comprised of:

Frontend App

The Web Interface is a React SPA built to modern standards with TypeScript, Vite, Tailwind and other popular packages from the React ecosystem. Webauthn APIs are used in-browser to create and manage Passkeys. tRPC is used as the client for the Infinex Backend, ensuring end-to-end type safety across the stack.

Backend Services

The backend is written in TypeScript and runs on Cloudflare Workers, which provide a secure and scalable platform for managing user sessions, processing backend logic, handling database queries, reading and caching blockchain data, and orchestrating onchain transactions.

The backend stores persistent state in Cloudflare Durable Objects and a MySQL database hosted by PlanetScale.

Onchain Infrastructure

The Infinex Account is implemented for EVM chains as Smart Contracts written in Solidity, and as Solana Programs written in Rust. Although there are similarities between the Infinex Smart Contract design and ERC-4337 (Account Abstraction) implementations, Infinex’s entire onchain implementation is custom and has been through multiple independent audits.

Gelato is used to pay gas and relay transactions from the backend services to EVM chains. Solana transactions are sent directly to Infinex’s RPC providers, Triton and Helius.

Protocol Integrations

Interacting with any onchain program, contracts, or protocols carries inherent risks that Infinex has solved by developing a new architecture. Integrations are isolated to limit the exposure of the Infinex Account while also providing the user with granular opt-in to each onchain protocols, as well as reducing the complexity and upgrade frequency of the main Infinex Account programs and contracts.

In Infinex, integrations are programs and contracts that the main Infinex Account interacts with through a limited interface, providing a degree of separation from the keys that operate that Account. Infinex has developed an Integration SDK, which can be used by the Infinex team and other protocols. Integrations must be approved by the Infinex Council and then enabled by users on an opt-in basis. They can then be independently updated to provide new functionality.

Protocol integrations are entirely onchain; Infinex maintains the backend and user interfaces to ensure a seamless user experience. This modular architecture allows Infinex to scale its feature set beyond that of centralized exchanges by leveraging onchain protocols, taking advantage of composability while minimising complexity. The main Account implementation can be updated infrequently and carefully; while integrations can be updated regularly, with each discretely opted into by users.

Security Overview

Passkeys

Passkeys are the core mechanism used to secure Infinex Accounts and authorise onchain transactions. When a user creates an Infinex Account, a corresponding passkey (a public-private key pair) is created by their operating system or password manager and stored securely on their device. At no point is the private key accessible to the browser or Infinex, and unlocking the private key to sign a challenge involves an on-device biometric or password challenge (depending on the operating system or password manager used).

Passkeys are (a) resilient to phishing, (b) always strong, and (c) designed so that there are no shared secrets. The private key cannot be written down, forgotten, or used to sign challenges on non-Infinex domains. This makes them superior to both username/password authentication, and user self-custody of private keys as in traditional crypto wallets.

When users sign in to the Infinex App, the Backend service validates the passkey signature and creates a session to identify the user.

Because passkey signatures cannot currently be validated natively onchain, Turnkey is used to translate passkey signatures to chain-supported signatures (ECDSA for EVM chains, Ed25519 for Solana).

Turnkey

Turnkey provides non-custodial wallet infrastructure that securely holds public-private key pairs and can sign transactions on behalf of users. The private key can only be used to sign transactions when unlocked by the user’s passkey and is not accessible to either Infinex or Turnkey.

More information about Turnkey’s architecture, security design and practices can be found here.

Secure by design

Infinex has been designed to provide users with a combination of centralised exchange security (including account recovery, two-factor authentication and session elevation) and onchain security (including transparency, censorship resistance, and decentralisation).

This includes recognising that there is no “one size fits all” approach to security postures; users will have different levels of technical sophistication. The best security system is one that works and is also convenient and designed with likely usage behaviours in mind.

Funds Recovery

An example of secure design in practice is the fund recovery feature. If a user loses access to their Passkey, it is critical they retain a safe way to recover their funds from their Infinex Account.

This brings up two important questions:

How can the user authenticate themselves without their passkey?
Where should the funds be recovered?

Some users will be familiar with onchain security practices and have their own wallets. For these users, Infinex offers the ability to sign a transaction to recover funds using their own EOA.

For other users, social sign-on is more appropriate. Infinex supports both Apple and Google accounts for funds recovery, as both these providers offer high levels of security (including their own account recovery and 2FA features). Infinex acts as a trusted signer for these methods to translate off-chain verification of account authentication to signed funds recovery transactions.

Users can link all three recovery methods if desired to minimise the chance of being unable to complete funds recovery.

Infinex assumes both social and EOA recovery are less secure than passkey authentication:

Private keys can be stolen or phished
Social sign-on accounts may be stolen, phished, or reset through a forgotten password workflow.

To account for this, the user cannot specify their recovery address at the time of recovery. Instead, they are prompted to specify their recovery address (which may be their wallet or centralised exchange deposit address) before their Infinex Account deposit address is displayed.

By designing user experience and account security together, Infinex has developed a recovery mechanism that is uniquely secure and resilient to loss.

Censorship Resistance

Funds recovery also plays a role in censorship resistance. If either (a) a user’s Infinex Account access is disabled or (b) Infinex ceases to exist, the transaction to recover funds can be signed by the associated recovery EOA and executed directly on the blockchain.

Some features for funds recovery (including bridging, recovery address cross-chain sync, and social sign-on) still require the Infinex backend to orchestrate; to improve resilience in the future, Infinex plans to develop an open source recovery app hosted on GitHub as well as additional onchain functionality to allow funds recovery in any circumstance.

Governance

Infinex is committed to providing a credibly neutral governance framework. This is important to ensure broad ecosystem support for transitioning from centralized services to onchain platforms like Infinex. Historically, there has been significant tribalism in the crypto ecosystem, yet very few people currently interact with crypto daily. User growth and adoption are the goals of every project in the space; Infinex represents a unique opportunity to unite these different factions around a shared aim of onboarding millions of new users to a non-custodial platform that relies on onchain infrastructure.

Infinex plans to launch a governance token, the Patron NFT, representing governance power in the project. A gated sale is planned to ensure wide distribution of this governance token. To access this sale, users must have created an account and deposited funds into it during one of the two launch season campaigns, Speedrun the waitlist and Craterun. These two campaigns distributed passes and tickets, allowing users to mint NFTs during the sale.

Infinex is a DAO, and a representative council governs the project (bounded delegation), consisting of five elected seats and two appointed seats. Broadly speaking, there is a continuum of governance structures available to govern a DAO. Direct token voting is at one end of this spectrum, where each token represents one vote. In the middle is an unbounded delegate system; tokens can be delegated to any number of representatives, each of which then exerts a weighted vote based on the power delegated to them. At the other end of the continuum is a bounded delegate system, or a representative council; in this coordination scheme, there are a finite number of delegates, each with equal voting power.

There are many different trade-offs with these approaches; however, by combining a bounded delegate system and operational seats, a very powerful and flexible governance framework emerges. This system avoids diffusion of responsibility and governance fatigue by ensuring that key responsibilities are directly delegated to operational council members for which the role is their sole priority. After many years of governance iteration this system is one of the most stable and least easily captured that has been deployed.

Conclusion

The Infinex platform results from many technical achievements invented, developed and deployed over the last decade. While combining all of these elements into a cohesive platform is not trivial, the underlying technologies make Infinex possible. The tireless efforts of thousands of engineers, builders and investors to progress onchain infrastructure has made Infinex possible; we are incredibly thankful for those efforts. In this regard, Infinex is the culmination and realisation of a decade’s worth of indefatigable effort by everyone who believes crypto has the capacity to transform the world. And yet, without adopting onchain technologies within our own ecosystem, we cannot credibly make a claim to displace TradFi. Infinex represents a critical step in this transformation of coordination. Once we can confidently demonstrate that crypto operates on onchain rails, we will then finally be ready to transform how the rest of the world coordinates.